An overarching goal of our research is to understand how the base excision repair (BER) pathway maintains genomic integrity and mediates epigenetic regulation, and how deficiencies in BER impact human health. A major focus is to discover how DNA glycosylases, which initiate BER, find and excise damaged or modified forms of 5-methylcytosine (mC). The most abundant modified DNA base in nature, mC is critical for epigenetic regulation in plants and animals and for restriction modification in archaea and bacteria. However, cytosine methylation also poses a danger because mC deaminates to T, generating G/T mispairs and C to T mutations that threaten genomic and epigenetic integrity and causes human diseases including cancer. Countering this threat are three different types of glycosylases that excise T from G/T mispairs; TDG and MBD4 in mammals and MIG in archaea and bacteria. While most glycosylases excise bases that are foreign to DNA (e.g., uracil) these enzymes face the daunting task of removing thymine bases arising by mC deamination but not those in the vast background of A:T pairs or in polymerase-generated G/T mispairs. Because glycosylase action on undamaged DNA is mutagenic, the specificity of these G/T glycosylases is critical, yet it is poorly defined. The current paradigm holds that specificity involves recognition of the mismatched guanine. We will rigorously test this model and investigate other potential specificity factors, to define the mechanism of G/T glycosylase specificity. Our studies will reveal features of TDG and MBD4 that may account for inefficient repair of mC deamination, a potential cause of point mutations implicated in cancer and genetic disease. BER also functions in epigenetic regulation by serving to ?erase? mC through active DNA demethylation. An established pathway in vertebrates involves oxidation of mC by a TET enzyme to give three oxy-mC products (hmC, fC, caC), excision of fC or caC by TDG, and subsequent BER to yield unmodified C. Our studies will address major gaps in the understanding of this essential pathway, by defining how TDG recognizes and removes fC and caC and how it is recruited to sites of DNA demethylation. We are also interested in how post-translational modifications regulate BER, and our current focus is on determining how TDG is regulated by SUMO modification. TDG is sumoylated at a single site, and it has a SUMO-interacting motif (SIM) that binds SUMO domains, including an intramolecular SUMO. While TDG is considered a model for understanding how sumoylation can regulate enzyme activity, many fundamental questions remain. Our studies will reveal how sumoylation alters TDG activity and how the SIM mediates these effects. We will also define mechanisms of SUMO conjugation and deconjugation and learn how the SIM modulates these processes. An in vitro conjugation-deconjugation system will be used to test the paradigm that sumoylation of TDG is required to regulate its product release and ensure faithful completion of TDG-initiated BER. Results of these studies will inform how BER deficiencies impact human health and could suggest new therapeutic approaches for treating diseases including cancer.